netdev - Re: [PATCH wireless-next 18/20] ti: Use pr

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1339756931.24527.16.camel@joe2Laptop>
Date:	Fri, 15 Jun 2012 03:42:11 -0700
From:	Joe Perches <joe@...ches.com>
To:	Luciano Coelho <coelho@...com>
Cc:	linux-kernel@...r.kernel.org,
	"John W. Linville" <linville@...driver.com>,
	linux-wireless@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH wireless-next 18/20] ti: Use pr_<level>

On Fri, 2012-06-15 at 11:36 +0300, Luciano Coelho wrote:
> Hi Joe,

Hi Luciano.

> As Johannes already mentioned, I have to also ask why do you need to do
> this?

To use a more consistent logging style.
To minimize unnecessary changes when adding
similar drivers based on this code.

>  I see the point, slightly, with the wl1251 code, because it is
> still using printks, but with wlcore and friends, there's no functional
> change.

Using a consistent logging style minimizes defects.

> As mentioned above, this change here is acceptable, but removing the \n
> from the macro and adding it at the end of each and every call is
> useless and will just skew git-blame and make it harder to do rebase or
> apply patches because of conflicts.

In practice, the merging won't be much of an issue.
Neither will git blame.

> > -#define wl1271_warning(fmt, arg...) \
> > -	pr_warning(DRIVER_PREFIX "WARNING " fmt "\n", ##arg)
> > +#define wl1271_warn(fmt, arg...)				\
> > +	pr_warn(DRIVER_PREFIX "WARNING " fmt, ##arg)
> 
> Using pr_warn instead of pr_warning is pointless.  pr_warning is the
> main definition, actually, and pr_warn is just a "mirror".

I expect one day to reverse the definitions.

The kernel is moving, albeit slowly to prefer
pr_warn to mirror dev_warn, netdev_warn, etc.

Most of these uses will eventually become netdev_warn.
netdev_warn still uses a \n.

> Again, moving the \n from the macro definition and changing all the
> callers is an unnecessary, intrusive change.
> 
> We have wanted to change "wl1271" to "wl12xx" for a long time, but we
> don't do it exactly because this means we need to change the calls
> everywhere.

Nothing in ever _needed_, no child will go
hungry or fed here.  This is a consistency change.

There are a couple of instances where \n is
currently used unnecessarily that are fixed in
the conversion.

$ git grep -E  "\bwl12[0-9]+_[a-z]+\s*\(.*\\\\n\"" drivers/net/wireless/ti
drivers/net/wireless/ti/wl12xx/cmd.c:   wl1271_debug(DEBUG_CMD, "FEM autodetect: %s, manufacturer: %d\n",
drivers/net/wireless/ti/wl12xx/cmd.c:   wl1271_debug(DEBUG_CMD, "FEM autodetect: %s, manufacturer: %d\n",
drivers/net/wireless/ti/wlcore/main.c:          wl1271_error("Incorrect ampdu action id=%x\n", action);

People make these mistakes because the more
common and adjacent dev_<level> code _use_
a newline.  Using a consistent style minimzes
these defects.

cheers, Joe

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html